def test_func(names):
from easydev import MultiProcessing
t = MultiProcessing(verbose=False, maxcpu=8)
for name in names:
t.add_job(keggfunc, name)
t.run()
return t
def test_func(names):
from easydev import MultiProcessing
t = MultiProcessing(verbose=False, maxcpu=8)
for name in names:
t.add_job(keggfunc, name)
t.run()
return t
def multicore_analysis(anova, drugs, maxcpu=2):
"""Function used by :class:`ANOVA` to perform multiprocess analysis
:param anova: an instance of :class:`ANOVA`
:param list drugs: list of drugs to analyse
:param int maxcpu: number of CPU to use
:return: the instance itself with the individual_anova attribute filled
with all results
"""
t = MultiProcessing(maxcpu=maxcpu)
for i, drug in enumerate(drugs):
if drug not in anova.individual_anova.keys():
t.add_job(analyse_one_drug, anova, drug)
t.run()
# populate the ANOVA instance with the results
for this in t.results:
drug = this[0]
result = this[1]
anova.individual_anova[drug] = result
return anova
def multicore_anova(ic50, genomic_features, drug_decode=None, maxcpu=2,
sampling=0):
"""Using 4 cores, the entire analysis took 15 minutes using
4 CPUs (16 Oct 2015).
:param ic50: a filename or :class:`IC50` instance.
:return: the anova instance itself (not the results); see example below.
::
from gdsctool qq qs.anova import multicore
master = multicore(dataset, maxcpu=2)
results = master.anova_all()
from gdsctools import ANOVAReport()
report = ANOVAReport(master, results)
report.create_html_pages(0
.. warning:: experimental. Seems to work but sometimes hangs forever.
"""
print("experimental code to run the analysis with several cores")
print("May takes lots or resources and slow down your system")
t1 = time.time()
master = ANOVA(ic50, genomic_features=genomic_features,
drug_decode=drug_decode, low_memory=True)
master.sampling = sampling
drugs = master.ic50.drugIds
t = MultiProcessing(maxcpu=maxcpu)
# add all jobs (one per drug)
for i, drug in enumerate(drugs):
t.add_job(analyse_one_drug, master, drug)
t.run()
# populate the ANOVA instance with the results
for this in t.results:
drug = this[0]
result = this[1]
master.individual_anova[drug] = result
print("\nTook " + str(time.time() - t1) + "seconds.")
return master
def multicore_analysis(anova, drugs, maxcpu=2):
"""Function used by :class:`ANOVA` to perform multiprocess analysis
:param anova: an instance of :class:`ANOVA`
:param list drugs: list of drugs to analyse
:param int maxcpu: number of CPU to use
:return: the instance itself with the individual_anova attribute filled
with all results
"""
t = MultiProcessing(maxcpu=maxcpu)
for i, drug in enumerate(drugs):
if drug not in anova.individual_anova.keys():
t.add_job(analyse_one_drug, anova, drug)
t.run()
# populate the ANOVA instance with the results
for this in t.results:
drug = this[0]
result = this[1]
anova.individual_anova[drug] = result
return anova
def main(args=None):
from sequana import logger
if args is None:
args = sys.argv[:]
user_options = Options(prog="sequana")
# If --help or no options provided, show the help
if len(args) == 1:
user_options.parse_args(["prog", "--help"])
else:
options = user_options.parse_args(args[1:])
options.verbose = not options.quiet
if options.multiple is True:
from sequana.modules_report.multi_summary import MultiSummary
if options.glob:
sms = MultiSummary(output_filename="multi_summary.html",
pattern=options.glob, verbose=options.verbose)
else:
sms = MultiSummary(output_filename="multi_summary.html",
verbose=options.verbose)
sys.exit(0)
# We put the import here to make the --help faster
if options.file:
options.glob = options.file
from easydev import MultiProcessing
from sequana.snaketools import FileFactory
ff = FileFactory(options.glob)
assert len(set(ff.extensions)) == 1, "Input files must have the same extensions"
extension = ff.all_extensions[0]
logger.info("Found %s files:" % len(ff.realpaths))
for this in ff.realpaths:
logger.info(" - " + this)
mc = MultiProcessing(options.thread, progress=True)
if extension in ["fastq", "fastq.gz"]:
for filename in ff.realpaths:
mc.add_job(get_fastq_stats, filename, options.sample)
elif extension.endswith("bed"):
for filename in ff.realpaths:
mc.add_job(get_bed_stats, filename)
elif extension.endswith("bam"):
for filename in ff.realpaths:
mc.add_job(get_bam_stats, filename)
mc.run()
# For the BED file only
if extension.endswith("bed"):
results = []
for i, this in enumerate(ff.filenames):
df = mc.results[i]
df = pd.DataFrame(df)
df = df.T
df.index.name = this
df = df.reset_index()
df["filename"] = [this] * len(df)
results.append(df)
df = pd.concat(results).set_index("filename")
print(df)
return df
results = {}
for i, this in enumerate(ff.filenames):
if i == 0:
df = mc.results[0]
df.index.name = this
else:
other = mc.results[i]
other.index.name = this
df = df.append(other)
# For the bed files only
results[this] = mc.results[i]
# For FastQ only
try:df.index = ff.filenames
except:pass
print()
print(df)
return df
